Bug 94886 – Add an optimized version of copyLCharsFromUCharSource for ARM

• Component: Web Template Framework
• Status: RESOLVED FIXED

[Benjamin Poulain]

Some more SIMD :)

[Benjamin Poulain]

Created attachment 160303 [details]
Patch

[Benjamin Poulain]

I could not use intrinsics here because:
-I need the explicit alignment for speed.
-I need the auto increment for speed.
-IIRC the interleaved load intrinsics are not always available anyway.

[Build Bot]

Comment on attachment 160303 [details]
Patch

Attachment 160303 [details] did not pass mac-ews (mac):
Output: http://queues.webkit.org/results/13562967

[WebKit Review Bot]

Comment on attachment 160303 [details]
Patch

Attachment 160303 [details] did not pass chromium-ews (chromium-xvfb):
Output: http://queues.webkit.org/results/13570906

[Benjamin Poulain]

Created attachment 160310 [details]
Patch

[WebKit Review Bot]

Comment on attachment 160310 [details]
Patch

Attachment 160310 [details] did not pass chromium-ews (chromium-xvfb):
Output: http://queues.webkit.org/results/13568917

[Build Bot]

Comment on attachment 160310 [details]
Patch

Attachment 160310 [details] did not pass win-ews (win):
Output: http://queues.webkit.org/results/13559960

[Early Warning System Bot]

Comment on attachment 160310 [details]
Patch

Attachment 160310 [details] did not pass qt-ews (qt):
Output: http://queues.webkit.org/results/13559975

[Benjamin Poulain]

Comment on attachment 160310 [details]
Patch

I'll fix that tomorrow.

[Gyuyoung Kim]

Comment on attachment 160310 [details]
Patch

Attachment 160310 [details] did not pass efl-ews (efl):
Output: http://queues.webkit.org/results/13566937

[Early Warning System Bot]

Comment on attachment 160310 [details]
Patch

Attachment 160310 [details] did not pass qt-wk2-ews (qt):
Output: http://queues.webkit.org/results/13563945

[Peter Beverloo (cr-android ews)]

Comment on attachment 160310 [details]
Patch

Attachment 160310 [details] did not pass cr-android-ews (chromium-android):
Output: http://queues.webkit.org/results/13569880

[Benjamin Poulain]

Created attachment 160481 [details]
Patch

[Benjamin Poulain]

This new version improves the performance when we cannot use Neon, and is just as fast when the input is big enough.

[Gavin Barraclough]

Comment on attachment 160481 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=160481&action=review

This looks awesome, didn't know about the interleaved unpack, that's really handy.

> Source/WTF/wtf/text/ASCIIFastPath.h:134
> +#elif COMPILER(GCC) && CPU(ARM_NEON) && !(PLATFORM(BIG_ENDIAN) || PLATFORM(MIDDLE_ENDIAN))

Your optimized path skips an ASSERT to check the upper bits are zero; might be worth adding "&& defined(NDEBUG)" so that debug builds get the C-loop with the ASSERT.

> Source/WTF/wtf/text/ASCIIFastPath.h:141
> +        do {

I think WebKit coding style is no parens here.

[Benjamin Poulain]

Committed r133100: <http://trac.webkit.org/changeset/133100>